Depression is a major health problem, on a par with heart disease in its annual cost of $43.7 billion. Women, compared with men, have a 2 to 1 incidence of major depression and are prone to develop episodes at times of reproductive hormonal change. The focus of this revised application is premenstrual dysphoric disorder (PMDD), a depressive disorder in the DSM-1V. In a pilot study, we observed in PMDD, but not in normal control (NC) subjects, an abnormal direction and a decreased magnitude of phase-shift responses in melatonin offset time alter a morning bright light pulse. The abnormality occurred in the symptomatic luteal, but not the asymptomatic follicular, menstrual cycle phase. In this proposal, we aim to replicate and extend these findings to determine whether abnormal phase-shifts to light occur in PMDD at other times of day. We will test the hypothesis that abnormal phase-shift responses to light also occur in the late subjective evening (during a delay portion of the phase-response curve, PRC), and during the early subjective morning (during an advance portion of the PRC). To assess phase-shift responses, we will measure the effects of bright (about 6,000 lux), 3-hour (h) light pulses on the critical circadian parameters of rhythmic plasma melatonin secretion, sleep electroencephalogram (EEG), and core body temperature. Responses to light at 2 different times of day will be compared between the follicular and luteal phases in women with PMDD and in normal control subjects. Altered phase-shift responses may contribute to disturbances in internal temporal order, and thereby result in mood disorders in predisposed individuals. These investigations will provide further insight into the pathophysiology of PMDD and other depressive disorders in women, and serve as a basis for refining light treatment interventions in the future.